"Open the Pandora's Box of LLMs: Jailbreaking LLMs through Representation ..."

Tianlong Li, Xiaoqing Zheng, Xuanjing Huang (2024)

Details and statistics

DOI: 10.48550/ARXIV.2401.06824

access: open

type: Informal or Other Publication

metadata version: 2024-02-16

a service of  Schloss Dagstuhl - Leibniz Center for Informatics